Le thème principal de cette thèse est celui des liens macro financiers. J’ai couvert trois questions différentes liées à ce sujet. Dans le premier chapitre, Gaël Giraud et moi-même développons un modèle pour la zone euro en réponse à de nombreuses critiques des modèles de politiques avant le GFC et en mettant l’accent sur l’interaction entre le secteur financier et la macroéconomie. Les deuxième et troisième chapitres portent sur le comportement du secteur financier au lendemain de la crise financière mondiale et ses implications pour la macroéconomie. Le chapitre 2 examine la pratique de l’abstention à l’égard des emprunteurs en difficulté. La question finalement pertinente dans ce chapitre est de savoir dans quelle mesure ce comportement influe sur l’économie réelle. Enfin, le troisième chapitre met en lumière un épisode de manipulation sur les marchés des matières premières. Cette prétendue manipulation n’était apparemment possible qu’en raison de la position dominante des banques sur le marché pendant la période qui a précédé la crise et par la suite. En fin de compte, je quantifie les effets d’un tel comportement et apporte la preuve d’un changement structurel du marché manipulé au cours de la période de manipulation alléguée. Le premier chapitre exploite un ensemble de données au niveau des banques, tandis que dans les chapitres 2 et 3, je développe des modèles macroéconomiques structurels. En particulier, le modèle de système dynamique du deuxième chapitre est une innovation. Cette catégorie de modèles, et plus particulièrement un modèle de la taille que nous développons, n’a jamais été estimée et utilisée par la suite pour l’analyse des politiques. / The main theme of this thesis are macro-financial linkages. I covered three different questions related to this topic. In the first chapter Gaël Giraud and I develop a model for the Euro Area answering to many of the critiques of policy models before the Global Financial Crisis and with a focus on the interaction between the financial sector and the macroeconomy. The second and third chapter focus on behaviour of the financial sector in the aftermath of the Global Financial Crisis and its implications for the macroeconomy. Chapter 2 investigates the practice of forbearance towards stressed borrowers. The ultimately relevant question in this chapter is to what extend there is a feedback to the real economy due to this behaviour. Finally, the third chapter sheds light on an episode of manipulation in commodity markets. This alleged manipulation was apparently only possible due to the dominant market position banks took in the run up to the crisis and thereafter. Ultimately I quantify the effects of such behaviour and provide evidence of a structural change of the manipulated market during the period of alleged manipulation. The first chapter exploits a bank level dataset, whereas in chapter 2 and 3 I develop structural macroeconomic models. Especially the dynamical system model in the second chapter is an innovation. This class of models and more specifically a model of the size we develop has never been estimated and subsequently used for policy analysis.

How do financial variables, such as firms’ cash flow and banks’ capital, affect macroeconomic variables, such as investment and GDP growth? What are the macroeconomic effects of exchange rate depreciation in countries where firms and households have extensive foreign-currency liabilities? The doctoral thesis Empirical Essays on Macro-Financial Linkages consists of four separate papers in the field of empirical macroeconomics. The first three papers investigate the macroeconomic implications of financial-market imperfections. Imperfect information between borrowers and lenders makes it more costly for firms to finance investments with external funds than with internal funds. The external finance risk premium depends on the strength of firm balance sheets, which hence affects firm investment. The first paper, The Effect of Cash Flow on Investment: An Empirical Test of the Balance Sheet Channel, examines the importance of financial constraints for investment using a large Swedish firm-level data set which includes many smaller firms (where balance sheet effects are likely to be especially important). I find a positive effect of cash flow on investment, controlling for fundamental determinants of investment and any information in cash flow about investment opportunities. As predicted by the balance sheet channel, the estimated effect of cash flow on investment is especially large for firms which, a priori, are more likely to be financially constrained (low-dividend, small and non-group firms). Moreover, the investment-cash flow sensitivity is significantly larger and more persistent during the first half of the sample period, which includes a severe banking crisis and recession. The second paper, Credit Matters: Empirical Evidence on U.S. Macro-Financial Linkages, written jointly with Tamim Bayoumi, estimates the impact of an adverse shock to bank capital on credit availability and spending in the United States, allowing for feedback from spending and income through the balance sheets of banks, firms and households. We find that an exogenous fall in the bank capital/asset ratio by one percentage point reduces real GDP by some 1 ½ percent through its effects on credit availability, while an exogenous fall in demand of 1 percent of GDP is gradually magnified to around 2 percent through financial feedback effects. The third paper, The Effects of Real Exchange Rate Shocks in an Economy with Extreme Liability Dollarization, studies the effects of real exchange rate depreciation in Bolivia, where over 95 percent of bank credit is denominated in dollars. Currency depreciation increases the domestic-currency value of foreign-currency liabilities and the debt service burden, thus adversely affecting firm balance sheets. A key issue for policymakers in countries with widespread foreign-currency borrowing is whether depreciation would have the standard, expansionary effect on output, or if an adverse balance sheet would dominate. I find that real exchange depreciation has negligible effects on output, since a contractionary balance-sheet effect on investment is counteracted by the standard expansionary effect on net exports. The fourth paper, Uncovered Interest Parity in a Partially Dollarized Developing Country: Does UIP Hold in Bolivia? (And If Not, Why Not?), studies another aspect of macro-financial linkages. The so-called uncovered interest parity (UIP) condition states that interest rate differentials compensate for expected exchange rate changes, equalizing the expected returns from holding assets which only differ in terms of currency denomination. Because of data availability problems, there is a lack of empirical tests of UIP for developing countries. The paper studies the case of Bolivia, where there are bank accounts which only differ in terms of currency denomination (bolivianos or U.S. dollars). I find that UIP does not hold in Bolivia, but that the deviations are smaller than in most other studies of developed and emerging economies. / Diss. Stockholm : Handelshögskolan, 2009 Sammanfattning jämte 4 uppsatser